Employee Retention and Turnover

Why Employees Stay or Leave

Peter W. Hom, David G. Allen, Rodger W. Griffeth

• Publisher: Routledge
• Year: 2019
• ISBN:
  9781351382229
• Edition: 1
• Topic: Business
• Subtopic: Human Resource Management
• Index: Business

This exploration of what employee turnover is, why it happens, and what it means for companies and employees draws together contemporary and classic theories and research to present a well-rounded perspective on employee retention and turnover. The book uses models such as job embeddedness theory, proximal withdrawal states, and context-emergent turnover theory, as well as highlights cultural differences affecting global differences in turnover.

Employee Retention and Turnover contextualises the issue of turnover, its causes and its consequences, before discussing underrepresented antecedents of turnover, key aspects of retention and methods for regulating turnover, and future research directions.

Ideal for both academics and advanced students of industrial/organizational psychology, Employee Retention and Turnover is essential for understanding the past, present, and future of turnover and related research.

1

WHAT IS EMPLOYEE TURNOVER, WHY IS IT IMPORTANT, AND HOW IS IT MEASURED?

The Practical and Academic Significance of Turnover

Employee turnover—or voluntary termination of employees from employing organizations (Mobley, 1982)—has attracted interest from employers and scholars alike for a century (Hom, Lee, Shaw, & Hausknecht, 2017). As early as 1917, management consultants and academicians have written about turnover’s costs, its amelioration, and its causes (Hom et al., 2017). Managers have long focused on employee turnover because of the personnel costs incurred when employees quit, such as those for hiring and training replacements (Cascio, 2000). To illustrate, turnover costs can range from 90% to 200% of annual pay (Allen, Bryant, & Vardaman, 2010). Though difficult to compute for some occupational fields, turnover can impair productivity because short-staffed workforces may underproduce, while inexperienced replacements may produce inefficiently during early employment (Hom & Griffeth, 1995).

Beyond such familiar tangible expenses, other ramifications of high turnover may prove costly or worrisome (Hom & Griffeth, 1995). In particular, corporate America frets about excessive attrition among minority and female incumbents that can impede its progress toward a diversified staffing in executive posts or occupational fields where these demographic subgroups are underrepresented (Hom, Roberson, & Ellis, 2008). What is more, human capital loss can stymie how firms deliver products or services or even enter new markets. Thus, talent recruiters at Fortune 1000 firms report that unfilled STEM (science, technology, engineering, and mathematics) jobs resulted in “lower productivity” (56%) and “limits to business growth” (47%; Bayer Corporation, 2014). While the existence of a STEM shortage is widely debated, STEM attrition surely worsens the dysfunctional effects of insufficient STEM supply in firms (Mervis, 2014). Further, such talent losses most harm source firms when leavers join competing (destination) firms (and thus supply proprietary knowledge or former clients to them), or becoming new competitors (as start-ups; Campbell, Ganco, Franco, & Agarwal, 2012; Somaya, Williamson, & Lorinkova, 2008; Wezel, Cattani, & Pennings, 2006). Finally, turnover can endanger entire industries, such as manufacturing in export-oriented processing zones in developing countries (notably, China and Mexico). There, 60% to 100% turnover rates among Chinese and Mexican assemblers (Miller, Hom, & Gomez-Mejia, 2001; Qin, Hom, & Xu, 2019; West, 2004) can boost labor costs, product defects, and production shortfalls, and thus hamper global supply chains (dependent on finished goods) and export-driven growth of emerging economies (Jiang, Baker, & Frazier, 2009).

Although turnover remains a perennial concern for employers, it continues to be a lively and enduring fascination for scholarly inquiry. Though such investigations began 100 years ago (Hom et al., 2017), March and Simon (1958) legitimize this scholarly focus by theorizing that effective organizational functioning hinges on employee decisions to perform as well as their decisions to participate (vs. withdraw from organizations, aka turnover). Over 1,000 articles have appeared on this topic alone (Steers & Mowday, 1981). Employee turnover is a popular criterion for validating or extending general theories of motivation, such as expectancy theory (Hom, 1980), equity theory (Dittrich & Carrell, 1979), the theory of reasoned action (Hom & Hulin, 1981; Westaby, 2005), and the theory of planned behavior (Van Breukelen, Van der Vlist, & Steensma, 2004). Moreover, theories of organizational behavior often regard turnover as one among many outcomes of their motivational processes. To illustrate, models about job characteristics (Grant, 2007; Hackman & Oldham, 1980), leadership (Graen & Ginsburgh, 1977; Nishii & Mayer, 2009), influence tactics (Reina, Rogers, Peterson, Bryon, & Hom, 2018), and network closure (Hom & Xiao, 2011) explicate how turnover may arise from their central explanatory constructs.

Of course, employee turnover is a significant motivated behavior in its own right, inspiring theoretical formulations explicating its occurrence. Pioneering the first formal theory of turnover, March and Simon (1958) theorized that leaving hinges on movement desirability and ease. Drawing from this model, Price (1977; Price & Mueller, 1981) and Mobley (1977; Mobley, Griffeth, Hand, & Meglino, 1979) derived models that represent movement desire and ease with job satisfaction and job alternatives, respectively. Early—and successful—validations of the Price–Mobley models (Mobley, Horner, & Hollingsworth, 1978; Price & Mueller, 1981) inspired widespread empirical and theoretical critiques of these models (Hom, Caranikis-Walker, Prussia, & Griffeth, 1992; Price & Mueller, 1986). Over the next 15 years, other theorists began revising the first-generation Price–Mobley models by reconfiguring structural pathways (Hom & Kinicki, 2001) or elaborating subsystems (e.g., job search → quit path or dissatisfaction → quit path; Hulin, Roznowski, & Hachiya, 1985; Steel, 2002), expanding the array of causal determinants (Steers & Mowday, 1981), or reconceptualizing the turnover process (e.g., invoking turnover drivers besides job attitudes and specifying additional turnover paths; Lee & Mitchell, 1994). Breaking away from the March–Simon legacy, other scholars adapted turnover models from other research traditions, such as Rusbult and Farrell’s (1983) investment model (based on Kelley and Thibaut’s interdependence theory) and Sheridan and Abelson’s (1983) cusp catastrophe model. In a radical departure from historic focus on why people quit, Mitchell, Holtom, Lee, Sablynski, and Erez (2001) envisioned the forces on why people stay, introducing job embeddedness. Embeddedness scholarship has since dominated modern turnover research, demonstrating that this construct explains additional variance in turnover beyond that by standard turnover antecedents, but also other workplace behaviors (Lee, Burch, & Mitchell, 2014; Li, Lee, Mitchell, Hom, & Griffeth, 2016).

Rather than testing comprehensive or complex theories about turnover (the dominant concern of turnover scholarship) per se, other scholars scrutinize special turnover phenomena, such as the relationship between individual (or firm) performance and turnover (Shaw, 2011; Sturman, Shao, & Katz, 2012), turnover contagion (Felps et al., 2009; Shapiro, Hom, Shen, & Agarwal, 2016), the impact of job-satisfaction trajectories (Liu, Mitchell, Lee, Holtom, & Hinkin, 2012), the honeymoon-hangover effect (Wang, Hom, & Allen, 2017), commitment mindsets (Hom, Mitchell, Lee, & Griffeth, 2012; Xu & Payne, 2018), high-performance work practices (Hom et al., 2009), social network effects (Ballinger, Cross, & Holtom, 2016; Feeley & Barnett, 1997), employee guarding (Gardner, Munyon, Hom, & Griffeth, 2018), and resignation styles (Klotz & Bolino, 2016). In summary, employee turnover is a crucial organizational phenomenon that has commanded longstanding and pervasive attention from managers and scholars for over a century. While organizations worry about turnover’s varied dysfunctional effects, academicians regard turnover as a striking expression of employee malaise or organizational malfunction.

Definitions of Employee Turnover

Voluntary vs. Involuntary Turnover

Employee turnover is classically defined as an employee’s “voluntary” severance of his or her current employment ties (Mobley, 1982; Price, 1977). Scholars have focused on employees’ self-initiated leaving because such behavior represents the most appropriate criterion for validating theories of “individual motivated choice behavior” (Campion, 1991, p. 199), while employers are most concerned because they have less control over this form of leaving than involuntary terminations. To illustrate, prevailing theories of turnover presume that employees autonomously decide to terminate their employment because they find the job disagreeable or find other jobs more appealing (Hom & Griffeth, 1995). This standard definition construes voluntary quits as separations from employing organizations and excludes promotions, job transfers, or other internal job movements. Further, this notion typically excludes cessation of membership in organizations where members receive no payment, such as volunteers, interns, and parishioners (though turnover models have been generalized to explain their departures).

Given its intuitive appeal, many researchers and practitioners operationally measure voluntary turnover as a dichotomy, differentiating employee-initiated terminations from employer-initiated (e.g., dismissals, layoffs). Despite their simplicity, the types of leaving that constitute voluntary self-determined quits often vary across employers and scholars (Mobley et al., 1979), making determination of whether turnover is voluntary or not difficult. To illustrate, some researchers regard pregnancy as a form of voluntary exit (Lee, Mitchell, Wise, & Fireman, 1996; Marsh & Mannari, 1977), though others say otherwise (Mirvis & Lawler, 1977; Waters, Roach, & Waters, 1976). Moreover, changing laws and human resource management practices may prompt redefinition of turnover voluntariness. While traditionally construed as a form of involuntary turnover (Hanisch & Hulin, 1990), retirement increasingly resembles voluntary turnover as mandatory retirement is now outlawed and employees increasingly receive defined contribution (e.g., 401K) instead of defined benefits plans (enabling them to retire on their own timetable when they have accumulated enough pension funds; Hom, 2011).

Besides inconsistent classification, the measurement of the causes or reasons for leaving is also subject to various errors. Turnover research often relies on personnel files from employers who determine via exit interviews or supervisory observations to ascertain turnover motives. Yet exiting employees may not reveal truthfully their reasons for leaving to avoid “burning their bridges” (thus underreporting negative job events; Griffeth & Hom, 2001; Hom, 2011). Employers or supervisors too may distort official records, such as classifying firings as voluntary quits to safeguard leavers’ reputation (and avoid defamation lawsuits) or recording voluntary quits as layoffs to enable leavers to qualify for unemployment compensation (Hom & Griffeth, 1995). Indeed, Mobley et al. (1979) had long recognized that “personnel practitioners readily admit that a variety of factors influence the administratively recorded reason for attrition” (p. 515). Further, existing classification schemes may be deficient by failing to capture all exit reasons or recording only one reason when multiple ones underlie leaving (Campion, 1991).

Given that accurate determination of turnover voluntariness is crucial for its prediction and control, it behooves scholars and firms to give more attention to criterion measurement. Conceivably, the simplistic remedy is to classify all turnover cases as voluntary (regardless of reason, and thus obviating the need to assess leavers’ motives) if they are not officially initiated by employers. Such an approach fits with Maertz and Campion’s (1998) definition of voluntary turnover as “instances wherein management agrees that the employee had the physical opportunity to continue employment with the company, at the time of termination” (p. 50). Thus, “voluntariness means that there was no impediment to continued employment from physical disability or from company management” (p. 50). Under this system, all reasons other than dismissals, layoffs, or retirement would represent voluntary quits (e.g., quitting due to pregnancy or spousal relocation) as they imply “individual choice, even though the employee may feel as though the choice to stay is extremely costly” (p. 51).

All the same, scholars and employers often prefer fine-grained assessment of turnover voluntariness that includes assessment of quit motives to enable more effective turnover management and prediction (Hom & Griffeth, 1995; Mobley et al., 1979). To illustrate, determination that so-called “voluntary” quits are often driven by family reasons (e.g., leaving to care for dependents or follow a relocating spouse) than job dissatisfaction or perceived alternatives (standard turnover causes; Hom et al., 2017) suggests that prevailing turnover models can be enhanced by including predictors reflecting external influences (Price & Mueller, 1986) that can forecast such exits (such as work–family conflicts, anticipated spousal relocations, or family pressures to quit; Hom & Kinicki, 2001; Lee et al., 1996; Maertz & Campion, 2004). While exit interviews are common ways to assess quit motives (Griffeth & Hom, 2001), turnover scholars since Mobley et al. (1979) have long prescribed additional assessments for triangulation (Hom & Griffeth, 1995), such as delayed or third-party confidential surveys or interviews with leavers (Hom, 2011), given that leavers often misrepresent their motives to firm representatives at departure time (Campion, 1991; Hom et al., 2012). Attesting to the value of assessing employees’ perception over volitional control over leaving, Li et al. (2016) demonstrated that job attitudes and job embeddedness more accurately predict turnover among enthusiastic stayers and leavers (people who either want to—and can—stay or leave). By contrast, those antecedents poorly predicted turnover among reluctant stayers and leavers (employees perceiving little control over their decisions to stay or leave). These findings dispute the classic definition of voluntary quits as employee-initiated exits (or turnover not mandated by employers), revealing that such quits still vary in perceived or experienced voluntariness.

Although more valid and thorough assessment of turnover reasons may increase accuracy of classifying whether or not a turnover case is voluntary, Campion (1991) and Maertz and Campion (1998) noted that “mutual separations” (where both employees and employers agree to end their relationships), quittin...